A prognostic model for classification of COVID-19 severity based on clinical and laboratory testing data

This study aimed to build an machine learning based model to predict the COVID-19 severity and reveal risk factors related to COVID-19 severity based on laboratory testing and clinical data for 420 participants, using tree-based models such as XGBoost, LightGBM, random forest. We calculated the Odds Ratios (OR) to investigate whether the top-ranked features were statistically significant for severity classification, turning out that high sensitivity C-reactive protein (hs-CRP) was the most important feature for determining of COVID-19 severity and XGBoost model showed the highest performance in classifying COVID-19 severity and healthy controls with F1score (0.84) and AUC (0.87). We expect that our results are of considerable significance for early screening for diagnosing COVID-19 severity, which, in turn, assist in further retrospective research for uncommon infectious diseases. © 2023 IEEE.

Used Car Price Prediction in Surabaya Using Random Forest Regressor Algorithms

The use of private vehicles during the Covid-19 pandemic has increased because private vehicles, especially cars, are considered as the safest mode of transportation to maintain distance and prevent transmission of the Covid-19 virus. Based on data from two different Indonesian secondary car market place, a comparison of a price sample of Car X in the city of Surabaya with the specifications for the 2015 to 2018 car years with car milage under 1000 kilometers, the used cars have a variety of prices hence a used car price prediction system is needed so that people can find out the average price of used cars sold in the market. In this study the author will use the Random Forest Regressor as a machine learning algorithm to predict the price of a used car with a dataset from the AtapData website. The reason for choosing the Random Forest Regressor is because the algorithm has the power to handle large amounts of data with high dimensions with categorical and numerical data types. The evaluation method used in this study is the Root Mean Absolute Error which produces a value of 0.55612 for validation data and 0.56638 for testing data, while the evaluation proceed with Mean Absolute Error produces a value of 0.45208 for validation data and 0.47576 for testing data. © 2022 IEEE.

HLA Class I allele is associated with a lower risk to develop Post-COVID19 interstitial lung manifestations

Introduction: The HLA Class I genes codify crucial molecules in developing the immunological response against pathogenic agents such as SARS-CoV-2. We aimed to assess HLA-A alleles associated with COVID-19 subsequent pulmonary complications as interstitial lung manifestations (ILM). Material(s) and Method(s): 209 Mexican mestizo patients with a positive RT-PCR test for SARS-CoV-2 and confirmed clinical diagnosis of COVID-19 were included. The participants were monitored three months after the hospital discharge through tomography;They were divided into two groups, 1) patients who developed ILM post-COVID19 (n = 85) and 2) those patients without tomographic evidence of ILM (n = 124). The HLA-A locus was genotyped by endpoint PCR using Micro SSP Generic HLA Class I kits. The clinical and demographic variables were analyzed by SPSS software. The alleles and genotypes were analyzed by 2 x 2 contingency tables, the value of p was obtained by Yates' correction. Result(s): There is no significant difference in age, sex, BMI, hospitalization days, PAO2/FIO2, or invasive mechanical ventilation. The alleles HLA-A*02:01, *24:02, and *68:01 are the most frequent in both study groups, grouping more than 60% of the alleles identified. On the other hand, the frequency of the HLA-A*01:01 allele was decreased in the group with interstitial lung manifestations at 3 months of discharge, compared to the group without interstitial lung manifestations (p= 0.004, OR = 0.13, IC95% 0.03-0.58). There is no significant difference in the genotypic frequencies. Conclusion(s): Subjects carrying the HLA-A*01:01 allele have a lower risk of developing interstitial lung manifestations posterior from COVID-19.

C3a and C5b-9 levels differentially predict COVID-19 severity and mortality

Introduction: SARS-COV-2 infection may result in pneumonia leading to ARDS and ICU treatment. Activation of the complement system was verified in COVID-19 patients as a driving factor of thromboinflammation contributing to disease progression. Aim(s): To investigate C3a and C5b-9 levels as markers of COVID-19 severity and outcome. Method(s): 79 patients with a positive polymerase chain reaction (PCR) test for SARS-COV-2 were recruited;38 severe and 42 critical. Serum samples were collected on admission and analysed for C3a and C5b-9 levels by ELISA methodology. Patients were grouped into severe vs critical, non-intubated vs intubated and survivors vs nonsurvivors for comparisons. Statistical analysis by Mann-Whitney for non-parametric analysis and receiving operating curve (ROC) analysis was performed in GraphPad Prism. Result(s): A statistically significant increase for C3a and C5b-9 levels was observed between: a) severe vs critical (p<0.001 and p<0.0001), b) non-intubated vs intubated (p<0.001 and p<0.05) survivors vs non-survivors (p<0.001 and p<0.01). ROC analysis for ICU admission revealed a higher AUC for C5b-9 (0.771, p<0.001) compared to C3a (AUC= 0.686, p<0.01). A higher AUC was observed for C3a when analysis was performed for intubation need (AUC=0.746, p<0.001) or mortality (AUC=0.758, p<0.0001) compared to C5b-9 (intubation need AUC=0.663, p<0.05 and mortality AUC=0.637, p NS). Combining C3a and C5b9 revealed a powerful prediction tool for ICU admission (AUC=0.773, p<0.0001), intubation (AUC=0.756, p<0.0001) and mortality (AUC=0.753, p<0.001). Conclusion(s): C3a and C5b-9 may serve as prognostic tools either separately or in combination for the progression and outcome of COVID-19.

Predicting Happiness Score during Covid-19 Using Machine Learning

The COVID-19 pandemic has been a bad dream for many people. People suffered from job losses, leading to a low level of happiness. Happiness is the key to a healthy life, and predicting the happiness score of 156 countries will give the idea of a happiness index around the world during the COVID-19 pandemic. An open dataset of the happiness index has been picked from the World Happiness Report, which is manifested already in a United Nations conference. The available dataset splits into training data and testing data, respectively. The training data have fitted into different machine learning algorithms. After that, the prediction score has observed based on testing data. After applying a large number of algorithms, the highest accuracy of the resulting regression model is 97 percent. © 2022 IEEE.

Clarifying predictions for COVID-19 from testing data: The example of New York State.

With the spread of COVID-19 across the world, a large amount of data on reported cases has become available. We are studying here a potential bias induced by the daily number of tests which may be insufficient or vary over time. Indeed, tests are hard to produce at the early stage of the epidemic and can therefore be a limiting factor in the detection of cases. Such a limitation may have a strong impact on the reported cases data. Indeed, some cases may be missing from the official count because the number of tests was not sufficient on a given day. In this work, we propose a new differential equation epidemic model which uses the daily number of tests as an input. We obtain a good agreement between the model simulations and the reported cases data coming from the state of New York. We also explore the relationship between the dynamic of the number of tests and the dynamics of the cases. We obtain a good match between the data and the outcome of the model. Finally, by multiplying the number of tests by 2, 5, 10, and 100 we explore the consequences for the number of reported cases.